shepherd

Name: shepherd
Availability: InStock
Author: street-labs

Annotate source code and generate structured review prompts for AI coding agents.

Updated Feb 21, 2026

Search Tags:#code review #macOS #AI feedback #structured prompt #source code annotation

Ready to run

npx skills add https://github.com/street-labs/shepherd --skill shepherd-street-labs

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: shepherd
Source: https://github.com/street-labs/shepherd/tree/main/.config/opencode/skills/shepherd
Command: npx skills add https://github.com/street-labs/shepherd --skill shepherd-street-labs

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

The Shepherd Skill solves the challenge of providing precise feedback on AI coding agent work, allowing users to easily annotate source code and generate structured review prompts for precise instructions.

Core Features & Use Cases

Inline Code Annotating: Allows users to directly annotate code files for feedback.
Review Prompt Generation: Outputs a structured review prompt ready for use by the AI agent.
Native macOS App: Ensures code reviews are performed on the local machine, keeping data secure.

Quick Start

Annotate and review code in Shepherd by using the command: /shepherd <filepath>

Frequently Asked Questions about shepherd

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I generate structured review prompts for AI coding agents from source code?▼

You generate structured review prompts by directly annotating source code files, which the system then processes to output precise instructions for AI coding agents. This ensures your feedback is accurately translated into actionable tasks.

Does this code review tool work natively on macOS?▼

Yes, this code review tool is a native macOS application, ensuring that your source code annotation and AI feedback generation are performed locally on your machine. This setup keeps your codebase and review data secure on your local environment.

How do I annotate and review code files using the /shepherd command?▼

To annotate and review code, execute the command /shepherd <filepath> in your environment. This passes your target source code file to the platform, allowing you to add inline annotations and generate the structured review prompt.

Do I need a specific repository setup to use structured prompt generation for code reviews?▼

Yes, structured prompt generation requires a Shepherd repository with the appropriate launch script to be present in your environment. This repository provides the necessary foundation for the annotation platform to function correctly.

What is the best way to provide precise feedback on AI coding agent work locally?▼

The best way to provide precise feedback is to use inline code annotating within a native macOS app to mark up files directly. This approach generates structured review prompts that guide the AI agent with exact, contextual corrections.